We have various everyday measures for identifying the presence of consciousness, such as the capacity for verbal report and the intentional control of behavior. However, there are many contexts in which these measures are difficult to apply, and even when they can be applied one might have doubts as to their validity in determining the presence/absence of consciousness. Everyday measures for identifying consciousness are particularly problematic when it comes to ‘challenging cases’—human infants, people with brain damage, nonhuman (...) animals, and AI systems. There is a pressing need to identify measures of consciousness that can be applied to challenging cases. This paper explores one of the most promising strategies for identifying and validating such measures—the natural-kind strategy. The paper is in two broad parts. Part I introduces the natural-kind strategy, and contrasts it with other influential approaches in the field. Part II considers a number of objections to the approach, arguing that none succeeds. (shrink)

1. 1.1. Suffering is bad. This is why, ceteris paribus, there are strong moral reasons to prevent suffering. Moreover, typically, those moral reasons are stronger when the amount of suffering at st...

Explainable artificial intelligence (XAI) is an emerging, multidisciplinary field of research that seeks to develop methods and tools for making AI systems more explainable or interpretable. XAI researchers increasingly recognise explainability as a context-, audience- and purpose-sensitive phenomenon, rather than a single well-defined property that can be directly measured and optimised. However, since there is currently no overarching definition of explainability, this poses a risk of miscommunication between the many different researchers within this multidisciplinary space. This is the problem we (...) seek to address in this paper. We outline a framework, called Explanatory Pragmatism, which we argue has two attractive features. First, it allows us to conceptualise explainability in explicitly context-, audience- and purpose-relative terms, while retaining a unified underlying definition of explainability. Second, it makes visible any normative disagreements that may underpin conflicting claims about explainability regarding the purposes for which explanations are sought. Third, it allows us to distinguish several dimensions of AI explainability. We illustrate this framework by applying it to a case study involving a machine learning model for predicting whether patients suffering disorders of consciousness were likely to recover consciousness. (shrink)

Our society is in the middle of the AI revolution. We discuss several applications of AI, in particular medical causality, where deep-learning neural networks screen through big data bases, extracting associations between a patient’s condition and possible causes. While beneficial in medicine, several questionable AI trading strategies have emerged in finance. Though advantages in many aspects of our lives, serious threats of AI exist. We suggest several regulatory measures to reduce these threats. We further discuss whether ‘full AI robots’ should (...) be programmed with a virtual consciousness and conscience. While this would reduce AI threats via motivational control, other threats such as the desire for AI—human socioeconomic equality could prove detrimental. (shrink)

Could computers ever be conscious? Will they ever have ideas that one could attribute to them and not to the programmer? Will robots be able to 'feel pain', instead of processing bits from sensors informing about danger? Will they have true emotions? These questions may never be answered, but it makes sense to ask whether humans will ever attribute mind to artifacts. This paper suggests introducing a third level of claims regarding artificial intelligence (AI), which is in between 'weak AI' (...) and 'strong AI', the so-called 'attributed AI'. This level requires more than weak AI ('behave as if', which could be said of any desktop calculator), but is less presumptuous than strong AI ('computers that think', a claim that is hard to prove). Attributed AI can be measured. This paper discusses behavioral paradigms for measurements of attributed AI and presents first experimental data. (shrink)

Kammerer and Frankish (this issue) propose an exciting new research programme on the computational form of introspective systems. Pursuing this goal requires measures that can isolate introspective capacity from response biases and first-order processes. I suggest that metacognitive psychophysics is well placed to meet this challenge, allowing the mapping of introspective architectures in humans, animals, and artificial systems.

Consciousness is now a hot topic in both philosophy and the cognitive sciences, yet there is much controversy over how to measure it. First, it is not clear whether biased subjective reports should be taken as adequate for measuring consciousness, or if more objective measures are required. Ways to benefit from the advantages of both these measures in the form of ‘Type 2’ metacognitive measures are under development, but face criticism. Research into neurophysiological measures of consciousness (...) is potentially very valuable, but often hindered by its dependence on behavioural measures that are themselves controversial. Newer informational measures are theoretically interesting, but as yet not easily applicable, while other measures of consciousness are designed for application in clinical cases . Although much optimism accompanies research into measures of consciousness, the controversies and methodological problems reviewed here suggest that making progress in this area may be more difficult than is often acknowledged. (shrink)

This edited volume presents perspectives from computer science, information theory, neuroscience and brain imaging, aesthetics, social sciences, psychiatry, and philosophy to answer frontier questions related to artificial intelligence and human experience. Can a machine think, believe, aspire and be purposeful as a human? What is the place in the machine world for hope, meaning and transformative enlightenment that inspires human existence? How, or are, the minds of machines different from that of humans and other species? These questions are responded to (...) along with questions in the intersection of health, intelligence and the brain. It highlights the place of consciousness by attempting to respond to questions with the help of fundamental reflections on human existence, its life-purposes and machine intelligence. The volume is a must-read for interdisciplinary and multidisciplinary researchers in humanities and social sciences and philosophy of science who wish to understand the future of AI and society. (shrink)

Sandberg et al. show that the Perceptual Awareness Scale scale is sensitive compared to confidence ratings and wagering in detecting accurate perception. They go on to argue that the PAS scale is hence a sensitive measure of conscious perception compared to confidence ratings, a claim disputed here. The fact that some visual content is conscious does not entail that the visual content relevant to making a discrimination is conscious. For example, if one saw a square but was only aware of (...) seeing a flash of something, then one has not consciously seen a square. When PAS and confidence ratings come in conflict, we suggest that it is confidence ratings that more reliably indicate the conscious status of contents allowing discrimination. (shrink)

Delirium has conventionally been considered a disorder of consciousness, but this remains a relatively unexamined precept. First, a review of the role of consciousness disruption in delirium is revised from an historical and diagnostic perspective. Second, consciousness measurement in routine assessment of delirium is considered. Conscious levels, comprising alertness and arousal, are most commonly used but are not representative of the multidimensionality of consciousness. Third, a justification for the exploration of phenomenal consciousness is presented. Three (...) candidate dimensions of phenomenal consciousness are identified as the pre-reflective state, phenomenal experience, and reflective thought. Finally, the clinical implications of a deeper understanding of delirium through measurement of phenomenal consciousness is considered. (shrink)

This article argues that if panpsychism is true, then there are grounds for thinking that digitally-based artificial intelligence may be incapable of having coherent macrophenomenal conscious experiences. Section 1 briefly surveys research indicating that neural function and phenomenal consciousness may be both analog in nature. We show that physical and phenomenal magnitudes—such as rates of neural firing and the phenomenally experienced loudness of sounds—appear to covary monotonically with the physical stimuli they represent, forming the basis for an analog relationship (...) between the three. Section 2 then argues that if this is true and micropsychism—the panpsychist view that phenomenal consciousness or its precursors exist at a microphysical level of reality—is also true, then human brains must somehow manipulate fundamental microphysical-phenomenal magnitudes in an analog manner that renders them phenomenally coherent at a macro level. However, Sect. 3 argues that because digital computation abstracts away from microphysical-phenomenal magnitudes—representing cognitive functions non-monotonically in terms of digits—digital computation may be inherently incapable of realizing coherent macroconscious experience. Thus, if panpsychism is true, digital AI may be incapable of achieving phenomenal coherence. Finally, Sect. 4 briefly examines our argument’s implications for Tononi’s Integrated Information Theory theory of consciousness, which we contend may need to be supplanted by a theory of macroconsciousness as analog microphysical-phenomenal information integration. (shrink)

In this chapter I offer two novel arguments for what I call strong primitivism about cognitive phenomenology, the thesis that there exists a phenomenology of cognition that is neither reducible to, nor dependent upon, sensory phenomenology. I then contend that strong primitivism implies that phenomenal consciousness does not require sensory processing. This latter contention has implications for the philosophy of artificial intelligence. For if sensory processing is not a necessary condition for phenomenal consciousness, then it plausibly follows that (...) AI consciousness (assuming that it is possible) does not require embodiment. The overarching goal of this paper is to show how different topics in the analytic philosophy of mind can be brought to bear on an important issue in the philosophy of artificial intelligence. (shrink)

The accurate measurement of the level of consciousness of a creature remains a major scientific challenge, nevertheless a number of new accounts that attempt to address this problem have been proposed recently. In this paper we analyze the principles of these new measures of consciousness along with other classical approaches focusing on their applicability to Machine Consciousness (MC). Furthermore, we propose a set of requirements of what we think a suitable measure for MC should be, discussing the (...) associated theoretical and practical issues. Using the proposed requirements as a framework for the design of an integrative measure of consciousness, we explore the possibility of designing such a measure in the context of current state of the art in consciousness studies. (shrink)

Dienes and Seth (2010) conclude that confidence ratings and post-decision wagering are two comparable and recommendable measures of conscious experience. In a recently submitted paper, we have however found that both methods are problematic and seem less suited to measure consciousness than a direct introspective measure. Here, we discuss the methodology and conclusions put forward by Dienes and Seth, and why we think the two experiments end up with so different recommendations.

Susan Schneider (2019) has proposed two new tests for consciousness in AI (artificial intelligence) systems, the AI Consciousness Test and the Chip Test. On their face, the two tests seem to have the virtue of proving satisfactory to a wide range of consciousness theorists holding divergent theoretical positions, rather than narrowly relying on the truth of any particular theory of consciousness. Unfortunately, both tests are undermined in having an ‘audience problem’: Those theorists with the kind of (...) architectural worries that motivate the need for such tests should, on similar grounds, doubt that the tests establish the existence of genuine consciousness in the AI in question. Nonetheless, the proposed tests constitute progress, as they could find use by some theorists holding fitting views about consciousness and perhaps in conjunction with other tests for AI consciousness. (shrink)

There has been an upsurge of interest lately in developing Wigner’s hypothesis that conscious observation causes collapse by exploring dynamical collapse models in which some purportedly quantifiable aspect of consciousness resist superposition. Kremnizer–Ranchin, Chalmers–McQueen and Okon–Sebastián have explored the idea that collapse may be associated with a numerical measure of consciousness. More recently, Chalmers–McQueen have argued that any single measure is inadequate because it will allow superpositions of distinct states of equal consciousness measure to persist. They suggest (...) a satisfactory model needs to associate collapse with a set of measures quantifying aspects of consciousness, such as the “Q-shapes” defined by Tononi et al. in their “integrated information theory” of consciousness. I argue here that Chalmers–McQueen’s argument against associating a single measure with collapse requires a precise symmetry between brain states associated with different experiences and thus does not apply to the only case where we have strong intuitions, namely human observers. In defence of Chalmers–McQueen’s stance, it might be argued that idealized artificial information processing networks could display such symmetries. However, I argue that the most natural form of any theory that postulates a map from network states to mind states is one that assigns identical mind states to isomorphic network states. This suggests that, if such a map exists, no familiar components of mind states, such as viewing different colours, or experiencing pleasure or pain, are likely to be related by symmetries. (shrink)

Consciousness can be measured in various ways, but different measures often yield different conclusions about the extent to which awareness relates to performance. Here, we compare five different subjective measures of awareness in the context of an artificial grammar learning task. Participants expressed their subjective awareness of rules using one of five different scales: confidence ratings , post-decision wagering , feeling of warmth , rule awareness , and continuous scale . All scales were equally sensitive to conscious knowledge. PDW, (...) however, was affected by risk aversion, and both RAS and SDS applied different minimal criteria for rule awareness. CR seems to capture the largest range of consciousness, but failed to indicate unconscious knowledge with the guessing criterion. We close by discussing the theoretical implications of scale sensitivity and propose that CR’s unique features enable a finer assessment of subjective states of awareness. (shrink)

SummaryEarly onset of menarche has been shown to be associated with breast cancer and ischaemic heart disease. Studies on age at menarche of the Malaysian population are poorly documented. This study aimed to determine the influence of anthropometric and socio-demographic factors on the age at menarche of university students in Malaysia. Data were obtained in 2010–11 from 961 students between the ages of 18 and 25 years from the University of Malaya using stratified sampling, and multiple regression analysis was applied. (...) Sixty-three per cent of students reached menarche at the age of 12 or 13 years, with the mean and median of age at menarche being 12.45±1.17 and 12.01 years, respectively. Menarcheal age was positively associated with height and negatively associated with BMI. Students from urban areas attained menarche earlier than those from rural areas. Students from small-sized families attained menarche earlier than those from larger families. First-born students experienced menarche earlier than those who were seventh-born or later. Obese and overweight students reached menarche earlier than students who were underweight or of normal weight. The variations in age at menarche among the Malaysian ethnic groups were statistically insignificant. The results suggest that heavier and first-born students from small families are more likely to attain menarche earlier than their counterparts. (shrink)

This article addresses two issues: the distinction between objective and subjective measures and the directness of such measures. It is argued that the distinction is unambiguous only when based on a methodological criterion rather than a semantic one. Different senses of directness are discussed: metaphysical, methodological, semantic, and causal.

Ultrasonic excitation has been widely used in the detection of microcracks on metal surfaces, but there are problems such as poor excitation effect of ultrasonic pulse, long time to reach the best excitation, and difficult to find microcracks. In this paper, an adaptive ultrasonic pulse excitation device and infrared thermal imaging technology have been combined, as well as their control method, to solve the problem. The adaptive ultrasonic pulse excitation device adds intelligent modules to realize automatic adjustment of detection parameters, (...) which can quickly obtain reliable excitation; the multidegree-of-freedom base realizes the three-dimensional direction change of the ultrasonic gun to adapt to different excitation occasions. When the appropriate ultrasonic excitation makes microcracks in the resonance state, the microcracks can be frictionated, which produce heat rise with the temperature. Then, the microcrack defect can be detected by the infrared thermal instrument through the different surface temperatures with imaging recognition method. Our detection experiments of the titanium alloy plates and the aluminum alloy profiles of marine engineering show that the method can get reliable detection parameters in a short time and measure the crack length effectively. It can be used in many aspects such as crack detection in mechanical structures or complex equipment operating conditions and industrial production processes. (shrink)

Civics and Moral Educationwas implemented as a new moral education programme in Singapore schools in 1992. This paper argues that the underlying theme is that of citizenship training and that new measures are under way to strengthen the capacity of the school system to transmit national values for economic and political socialisation. The motives and motivation for retaining a formal moral education programme have remained strong. A discussion of the structure and content of key modules in Civics and Moral Education (...) shows how curriculum writers have attempted to integrate separate strands of civics and moral education from earlier programmes and present them in a less divisive manner. The paper also assesses the impact of the written civics and moral education programme on teachers and pupils, and the way in which strategies have been used by curriculum agencies to overcome possible obstacles at the curriculum implementation phase. (shrink)

Edmund Husserl, the founder of phenomenological philosophy, developed the concept of the so-called pure transcendental consciousness. The author of the article asks whether the concept of consciousness understood this way can constitute a model for AI consciousness. It should be remembered that transcendental consciousness is the result of the use of the phenomenological method, the essence of which is referring to experience (“back to things themselves”). Therefore, one can legitimately ask whether the consciousness that AI (...) can achieve can possess the characteristics attributed by Husserl to pure transcendental consciousness. The answer to such questions seems to be negative because AI, as created by humans, can only operate in the field of phenomena. Human intelligence, however, is capable of operating at the ontological level. In the face of difficulties in understanding the phenomenon of consciousness on a scientific basis, the question arises about the possibility of using the phenomenological concept of consciousness developed by Husserl as a starting point in analyzes aimed at answering the question about AI consciousness (The aim of the article is not to discuss in detail the contemporary state of research on consciousness. Therefore, many results currently achieved mainly in the field of neuroscience are omitted. It is just about indicating the possible application of Husserl’s transcendental concept of consciousness in research on AI consciousness). (shrink)

This chapter discusses the main types of so-called ’subjective measures of consciousness’ used in current-day science of consciousness. After explaining the key worry about such measures, namely the problem of an ever-present response bias, I discuss the question of whether subjective measures of consciousness are introspective. I show that there is no clear answer to this question, as proponents of subjective measures do not employ a worked-out notion of subjective access. In turn, this makes the problem of (...) response bias less tractable than it might otherwise be. (shrink)

The hard problem of consciousness is to explain the experience of qualia. But everything gets easier once we realise that what has to be explained is not how qualia can exist as objective entities but rather why the conscious subject should believe that they exist. This essay lays out a programme for doing this. It makes radical proposals as to how the “qualia illusion” is created, and why sustaining this illusion is biologically adaptive.

The rise of mathematical developments in the theories of consciousness has led to new measures to detect consciousness in a system. The Integrated Information Theory is one of the best mathematical rooted attempts to quantify the level of consciousness in a system with Φ as the effective information generated in a system above its parts. Recently, the IIT has inspired the Perturbational Complexity Index to detect conscious states in patients with disorders of consciousness, and it has (...) shown to have almost perfect classification accuracy. In this study, we explore the statistical correspondence between the theoretical Φ and the experimental PCI through a neurocomputational model of coupled oscillators that can be artificially perturbed, which mainly focuses on the dynamics of collective synchronization between subsets of brain areas. Our results reveal that both measures are statistically related but, in principle, this relationship is far to be perfect. These results are discussed in the context of the model of coupled oscillators, which mainly focuses on the dynamics of collective synchronization between subsets of brain areas. (shrink)

Investigating the causal relationship between exposure and time-to-event outcome is an important topic in biomedical research. Previous literature has discussed the potential issues of using hazard ratio (HR) as the marginal causal effect measure due to noncollapsibility. In this article, we advocate using restricted mean survival time (RMST) difference as a marginal causal effect measure, which is collapsible and has a simple interpretation as the difference of area under survival curves over a certain time horizon. To address both measured and (...) unmeasured confounding, a matched design with sensitivity analysis is proposed. Matching is used to pair similar treated and untreated subjects together, which is generally more robust than outcome modeling due to potential misspecifications. Our propensity score matched RMST difference estimator is shown to be asymptotically unbiased, and the corresponding variance estimator is calculated by accounting for the correlation due to matching. Simulation studies also demonstrate that our method has adequate empirical performance and outperforms several competing methods used in practice. To assess the impact of unmeasured confounding, we develop a sensitivity analysis strategy by adapting the E-value approach to matched data. We apply the proposed method to the Atherosclerosis Risk in Communities Study (ARIC) to examine the causal effect of smoking on stroke-free survival. (shrink)

We propose four policies of ethical design of human-grade Artificial Intelligence. Two of our policies are precautionary. Given substantial uncertainty both about ethical theory and about the conditions under which AI would have conscious experiences, we should be cautious in our handling of cases where different moral theories or different theories of consciousness would produce very different ethical recommendations. Two of our policies concern respect and freedom. If we design AI that deserves moral consideration equivalent to that of human (...) beings, that AI should be designed with self-respect and with the freedom to explore values other than those we might impose. We are especially concerned about the temptation to create human-grade AI pre-installed with the desire to cheerfully sacrifice itself for its creators’ benefit. (shrink)

This paper addresses what we consider to be the most pressing challenge for the emerging science of consciousness: the measurement problem of consciousness. That is, by what methods can we determine the presence of and properties of consciousness? Most methods are currently developed through evaluation of the presence of consciousness in humans and here we argue that there are particular problems in application of these methods to nonhuman cases—what we call the indicator validity problem and the (...) extrapolation problem. The first is a problem with the application of indicators developed using the differences between conscious and unconscious processing in humans to the identification of other conscious vs. nonconscious organisms or systems. The second is a problem in extrapolating any indicators developed in humans or other organisms to artificial systems. However, while pressing ethical concerns add urgency to the attribution of consciousness and its attendant moral status to nonhuman animals and intelligent machines, we cannot wait for certainty and we advocate the use of a precautionary principle to avoid causing unintentional harm. We also intend that the considerations and limitations discussed in this paper can be used to further analyze and refine the methods of consciousness science with the hope that one day we may be able to solve the measurement problem of consciousness. (shrink)

Some patients, following brain injury, do not outwardly respond to spoken commands, yet show patterns of brain activity that indicate responsiveness. This is “cognitive-motor dissociation” (CMD). Recent research has used machine learning to diagnose CMD from electroencephalogram (EEG) recordings. These techniques have high false discovery rates, raising a serious problem of inductive risk. It is no solution to communicate the false discovery rates directly to the patient’s family, because this information may confuse, alarm and mislead. Instead, we need a procedure (...) for generating case-specific probabilistic assessments that can be communicated clearly. This article constructs a possible procedure with three key elements: (1) a shift from categorical “responding or not” assessments to degrees of evidence; (2) the use of patient-centred priors to convert degrees of evidence to probabilistic assessments; and (3) the use of standardized probability yardsticks to convey those assessments as clearly as possible. (shrink)

A wide range of problems of the relationship between consciousness and matter are discussed. Particular attention is paid to the analysis of the structure and properties of consciousness in the framework of information evolution. The role of specific (non-computational) properties of consciousness in the procedure of classical and quantum measurements is analyzed. In particular, the issue of "cloning" of consciousness (the possibility of copying its properties onto a new material carrier) is discussed in detail. We hope (...) that the generalized principle of complementarity formulated by us will open up new ways for studying the problems of consciousness within the framework of the fundamental physical picture of the world. (shrink)

In light of recent breakneck pace in machine learning, questions about whether near-future artificial systems might be conscious and possess moral status are increasingly pressing. This paper argues that as matters stand these debates lack any clear criteria for resolution via the science of consciousness. Instead, insofar as they are settled at all, it is likely to be via shifts in public attitudes brought about by the increasingly close relationships between humans and AI users. Section 1 of the paper (...) I briefly lays out the current state of the science of consciousness and its limitations insofar as these pertain to machine consciousness, and claims that there are no obvious consensus frameworks to inform public opinion on AI consciousness. Section 2 examines the rise of conversational chatbots or Social AI, and argues that in many cases, these elicit strong and sincere attributions of consciousness, mentality, and moral status from users, a trend likely to become more widespread. Section 3 presents an inconsistent triad for theories that attempt to link consciousness, behaviour, and moral status, noting that the trends in Social AI systems will likely make the inconsistency of these three premises more pressing. Finally, Section 4 presents some limited suggestions for how consciousness and AI research communities should respond to the gap between expert opinion and folk judgment. (shrink)

Typically, fair machine learning research focuses on a single decision maker and assumes that the underlying population is stationary. However, many of the critical domains motivating this work are characterized by competitive marketplaces with many decision makers. Realistically, we might expect only a subset of them to adopt any non-compulsory fairness-conscious policy, a situation that political philosophers call partial compliance. This possibility raises important questions: how does partial compliance and the consequent strategic behavior of decision subjects affect the allocation outcomes? (...) If k% of employers were to voluntarily adopt a fairness-promoting intervention, should we expect k% progress (in aggregate) towards the benefits of universal adoption, or will the dynamics of partial compliance wash out the hoped-for benefits? How might adopting a global (versus local) perspective impact the conclusions of an auditor? In this paper, we propose a simple model of an employment market, leveraging simulation as a tool to explore the impact of both interaction effects and incentive effects on outcomes and auditing metrics. Our key findings are that at equilibrium: (1) partial compliance by k% of employers can result in far less than proportional (k%) progress towards the full compliance outcomes; (2) the gap is more severe when fair employers match global (vs local) statistics; (3) choices of local vs global statistics can paint dramatically different pictures of the performance vis-a-vis fairness desiderata of compliant versus non-compliant employers; (4) partial compliance based on local parity measures can induce extreme segregation. Finally, we discuss implications for auditors and insights concerning the design of regulatory frameworks. (shrink)

We should not be creating conscious, humanoid agents but an entirely new sort of entity, rather like oracles, with no conscience, no fear of death, no distracting loves and hates.